Twitter Daily Digest

Security checks across malware telemetry and agentic risk

Overview

The skill is mostly a disclosed Twitter/X digest tool, but it can use local Notion credentials to create pages at a hard-coded default destination with broad trigger wording.

Review this skill carefully before installing. Use it only if you are comfortable with an authenticated Twitter/X CLI being used to fetch your followed accounts' recent posts and with digest files being left on disk. Before enabling Notion sync, change or verify the parent page, use a limited Notion integration token, and require an explicit user action for any Notion write.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (3)

Description-Behavior Mismatch

Medium
Confidence
88% confidence
Finding
The script includes a secondary capability to sync arbitrary Markdown into Notion and create pages, which goes beyond a Twitter digest collection skill. This expands the skill's effective privileges from local collection to outbound content publication, increasing the risk of unintended data exfiltration or misuse in an agent context where users may not expect external writes.

Context-Inappropriate Capability

Medium
Confidence
94% confidence
Finding
The code reads a local Notion API key from the user's home directory and uses it to perform authenticated API requests to create pages. In an agent environment, silently consuming local credentials and transmitting generated content to an external service creates a meaningful exfiltration and privilege-boundary risk, especially because the skill's stated purpose is Twitter digesting rather than cross-service publishing.

Vague Triggers

Medium
Confidence
80% confidence
Finding
The trigger phrases include broad everyday expressions such as '帮我看看推特' and '刷推', which can cause the skill to activate in contexts where the user did not intend bulk fetching, file creation, or downstream syncing behavior. In a skill with shell, filesystem, and network capabilities, accidental activation increases the risk of unintended data access and external requests.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal